Cs 430 info 430 information retrieval
This presentation is the property of its rightful owner.
Sponsored Links
1 / 47

CS 430 / INFO 430 Information Retrieval PowerPoint PPT Presentation


  • 96 Views
  • Uploaded on
  • Presentation posted in: General

CS 430 / INFO 430 Information Retrieval. Lecture 11 Latent Semantic Indexing Extending the Boolean Model. Course Administration. Assignment 1 If you have questions about your grading, send me email.

Download Presentation

CS 430 / INFO 430 Information Retrieval

An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Presentation Transcript


Cs 430 info 430 information retrieval

CS 430 / INFO 430 Information Retrieval

Lecture 11

Latent Semantic Indexing

Extending the Boolean Model


Course administration

Course Administration

Assignment 1

If you have questions about your grading, send me email.

The following are reasonable requests: the wrong files were graded, points were added up wrongly, comments are unclear, etc.

We are not prepared to argue over details of judgment.

If you ask for a regrade, the final grade may be lower than the original!


Course administration1

Course Administration

Assignment 2

The assignment has been posted.

The test data is being checked. Look for changes before Saturday evening.


Course administration2

Course Administration

Midterm Examination

Wednesday, October 14, 7:30 to 9:00 p.m., Upson B17. Open book.

Laptop computers may be used for lecture slides, notes, readings, etc., but no network connections during the examination.

A sample examination and discussion of the solution will

be posted to the Web site.


Cs 430 info 430 information retrieval1

CS 430 / INFO 430 Information Retrieval

Latent Semantic Indexing


Latent semantic indexing

Latent Semantic Indexing

Objective

Replace indexes that use sets of index terms by indexes that use concepts.

Approach

Map the term vector space into a lower dimensional space, using singular value decomposition.

Each dimension in the new space corresponds to a latent concept in the original data.


Deficiencies with conventional automatic indexing

Deficiencies with Conventional Automatic Indexing

Synonymy: Various words and phrases refer to the same concept (lowers recall).

Polysemy: Individual words have more than one meaning (lowers precision)

Independence: No significance is given to two terms that frequently appear together


Example

Example

Query: "IDF in computer-based information look-up"

Index terms for a document:access, document, retrieval,

indexing

How can we recognize that informationlook-up is related to retrieval and indexing?

Conversely, if information has many different contexts in the set of documents, how can we discover that it is an unhelpful term for retrieval?


Technical memo example titles

Technical Memo Example: Titles

 c1Human machine interface for Lab ABC computer applications

 c2A survey of user opinion of computer system response time

 c3The EPS user interface management system

 c4System and humansystem engineering testing of EPS

 c5Relation of user-perceived responsetime to error measurement

m1The generation of random, binary, unordered trees

m2The intersection graph of paths in trees

m3Graph minors IV: Widths of trees and well-quasi-ordering

 m4Graph minors: A survey


Technical memo example terms and documents

Technical Memo Example: Terms and Documents

TermsDocuments

c1c2c3c4c5m1m2m3m4

human100100000

interface101000000

computer110000000

user011010000

system011200000

response010010000

time010010000

EPS001100000

survey010000001

trees000001110

graph000000111

minors000000011


Technical memo example query

Technical Memo Example: Query

Query:

Find documents relevant to "human computer interaction"

Simple Term Matching:

Matches c1, c2, and c4

Misses c3 and c5


The index term vector space

The index term vector space

t3

The space has as many dimensions as there are terms in the word list.

d1

d2

t2

t1


Models of semantic similarity

Models of Semantic Similarity

Proximity models: Put similar items together in some space or

structure

•Clustering (hierarchical, partition, overlapping). Documents are considered close to the extent that they contain the same terms. Most then arrange the documents into a hierarchy based on distances between documents. [Covered later in course.]

•Factor analysis based on matrix of similarities between documents (single mode).

•Two-mode proximity methods. Start with rectangular matrix and construct explicit representations of both row and column objects.


Selection of two mode factor analysis

Selection of Two-mode Factor Analysis

Additional criterion:

Computationally efficient O(N2k3)

N is number of terms plus documents

k is number of dimensions


Figure 1

Figure 1

• term

document

query

--- cosine > 0.9


Mathematical concepts

Mathematical concepts

Singular Value Decomposition

Define X as the term-document matrix, with t rows (number of index terms) and d columns (number of documents).

There exist matrices T, S and D', such that:

X = T0S0D0'

T0 and D0 are the matrices of left and right singular vectors

T0 and D0 have orthonormal columns

S0 is the diagonal matrix of singular values


Dimensions of matrices

Dimensions of matrices

t x d

t x m

m x m

m x d

S0

D0'

X

=

T0

m is the rank of X< min(t, d)


Reduced rank

Reduced Rank

~

~

Diagonal elements of S0 are positive and decreasing in magnitude. Keep the first k and set the others to zero.

Delete the zero rows and columns of S0 and the corresponding rows and columns of T0 and D0. This gives:

X X = TSD'

Interpretation

If value of k is selected well, expectation is that X retains the semantic information from X, but eliminates noise from synonymy,and recognizes dependence.

^

^


Selection of singular values

Selection of singular values

t x d

t x k

k x k

k x d

S

D'

^

=

X

T

k is the number of singular values chosen to represent the concepts in the set of documents.

Usually, k« m.


Comparing two terms

Comparing Two Terms

^

The dot product of two rows of X reflects the extent to which two terms have a similar pattern of occurrences.

^

^

XX' = TSD'(TSD')'

= TSD'DS'T'

=TSS'T Since D is orthonormal

= TS(TS)'

To calculate thei, jcell, take the dot product between the i and j rows ofTS

Since S is diagonal, TS differs from T only by stretching the coordinate system


Comparing two documents

Comparing Two Documents

^

The dot product of two columns of X reflects the extent to which two columns have a similar pattern of occurrences.

^

^

X'X = (TSD')'TSD'

= DS(DS)'

To calculate thei, jcell, take the dot product between the i and j columns ofDS.

Since S is diagonal DS differs from D only by stretching the coordinate system


Comparing a term and a document

Comparing a Term and a Document

Comparison between a term and a document is the value of an individual cell of X.

X = TSD'

= TS(DS)'

where S is a diagonal matrix whose values are the square root of the corresponding elements of S.

^

^

-

-

-


Technical memo example query1

Technical Memo Example: Query

Terms Query

xq

human1

interface0

computer0

user0

system1

response0

time0

EPS0

survey0

trees1

graph0

minors0

Query:

"humansystem interactions on trees"

In term-document space, a query is represented by xq, a t x 1 vector.

In concept space, a query is represented by dq, a 1 x k vector.


Comparing a query and a document

Comparing a Query and a Document

A query can be expressed as a vector in the term-document vector space xq.

xqi= 1 if term i is in the query and 0 otherwise.

Let pqj be the inner product of the queryxqwith document dj in the term-document vector space.

pqj is the jth element in the product of xq'X.

^


Comparing a query and a document1

Comparing a Query and a Document

^

X

[pq1... pqj ... pqt] = [xq1 xq2 ... xqt]

document dj is column j of X

^

inner product of query q with document dj

query

^

pq' = xq'X

= xq'TSD'

= xq'T(DS)'

similarity(q, dj) =

cosine of angle is inner product divided by lengths of vectors

pqj

|xq| |dj|

Revised October 6, 2004


Comparing a query and a document2

Comparing a Query and a Document

In the reading, the authors treat the query as a pseudo-document in the concept space dq:

dq = xq'TS-1

To compare a query against document j, they extend the method used to compare document i with document j.

Take the jth element of the product of:

dqS and(DS)'

This is the jth element of product of:

xq'T (DS)' which is the same expression as before.

Note that dq is a row vector.

Revised October 6, 2004


Experimental results

Experimental Results

Deerwester, et al. tried latent semantic indexing on two test collections, MED and CISI, where queries and relevant judgments available.

Documents were full text of title and abstract.

Stop list of 439 words (SMART); no stemming, etc.

Comparison with:

(a) simple term matching, (b) SMART, (c) Voorhees method.


Experimental results 100 factors

Experimental Results: 100 Factors


Experimental results number of factors

Experimental Results: Number of Factors


Cs 430 info 430 information retrieval2

CS 430 / INFO 430 Information Retrieval

Extending the Boolean Model


Boolean diagram

Boolean Diagram

not (A or B)

A and B

A

B

A or B


Problems with the boolean model

Problems with the Boolean model

Counter-intuitive results:

Query q = A and B and C and D and E

Document d has terms A, B, C and D, but not E

Intuitively, d is quite a good match for q, but it is rejected by the Boolean model.

Query q = A or B or C or D or E

Document d1 has terms A, B, C,D and E

Document d2 has term A, but not B, C,D or E

Intuitively, d1 is a much better match than d2, but the Boolean model ranks them as equal.


Problems with the boolean model continued

Problems with the Boolean model (continued)

Boolean is all or nothing

•Boolean model has no way to rank documents.

•Boolean model allows for no uncertainty in assigning index terms to documents.

•The Boolean model has no provision for adjusting the importance of query terms.


Boolean model as sets

Boolean model as sets

d is either in the set A or not in A.

d

A


Extending the boolean model

Extending the Boolean model

Term weighting

•Give weights to terms in documents and/or queries.

•Combine standard Boolean retrieval with vector ranking of results

Fuzzy sets

•Relax the boundaries of the sets used in Boolean retrieval


Ranking methods in boolean systems

Ranking methods in Boolean systems

SIRE (Syracuse Information Retrieval Experiment)

Term weights

•Add term weights to documents

Weights calculated by the standard method of

term frequency * inverse document frequency.

Ranking

• Calculate results set by standard Boolean methods

• Rank results by vector distances


Relevance feedback in sire

Relevance feedback in SIRE

SIRE (Syracuse Information Retrieval Experiment)

Relevance feedback is particularly important with Boolean

retrieval because it allow the results set to be expanded

• Results set is created by standard Boolean retrieval

• User selects one document from results set

• Other documents in collection are ranked by vector

distance from this document


Boolean model as fuzzy sets

Boolean model as fuzzy sets

d is more or less in A.

d

A


Basic concept

Basic concept

•A document has a term weight associated with each index term. The term weight measures the degree to which that term characterizes the document.

•Term weights are in the range [0, 1]. (In the standard Boolean model all weights are either 0 or 1.)

•For a given query, calculate the similarity between the query and each document in the collection.

•This calculation is needed for every document that has a non-zero weight for any of the terms in the query.


Mmm mixed min and max model

MMM: Mixed Min and Max model

Fuzzy set theory

dAis the degree of membership of an element to set A

intersection (and)

dAB = min(dA, dB)

union (or)

dAB = max(dA, dB)


Mmm mixed min and max model1

MMM: Mixed Min and Max model

Fuzzy set theory example

standard fuzzy

set theory set theory

dA11000.50.500

dB10100.700.70

and dAB10000.5000

or dAB11100.70.50.70


Mmm mixed min and max model2

MMM: Mixed Min and Max model

Terms: A1, A2, . . . , An

DocumentD, with index-term weights: dA1, dA2, . . . , dAn

Qor = (A1or A2or . . . or An)

Query-document similarity:

S(Qor, D) = Cor1 * max(dA1, dA2,.. , dAn) + Cor2 * min(dA1, dA2,.. , dAn)

where Cor1 + Cor2 = 1


Mmm mixed min and max model3

MMM: Mixed Min and Max model

Terms: A1, A2, . . . , An

DocumentD, with index-term weights: dA1, dA2, . . . , dAn

Qand = (A1and A2and . . . and An)

Query-document similarity:

S(Qand, D) = Cand1 * min(dA1,.. , dAn) + Cand2 * max(dA1,.. , dAn)

where Cand1 + Cand2 = 1


Mmm mixed min and max model4

MMM: Mixed Min and Max model

Experimental values:

Cand1 in range [0.5, 0.8]

Cor1 > 0.2

Computational cost is low. Retrieval performance much improved.


Other models

Other Models

Paice model

The MMM model considers only the maximum and minimum document weights. The Paice model takes into account all of the document weights. Computational cost is higher than MMM.

P-norm model

DocumentD, with term weights: dA1, dA2, . . . , dAn

Query terms are given weights, a1, a2, . . . ,an

Operators have coefficients that indicate degree of strictness

Query-document similarity is calculated by considering each document and query as a point in n space.


Test data

Test data

CISICACMINSPEC

P-norm79106210

Paice77104206

MMM68109195

Percentage improvement over standard Boolean model (average best precision)

Lee and Fox, 1988


Reading

Reading

E. Fox, S. Betrabet, M. Koushik, W. Lee, Extended Boolean Models, Frake, Chapter 15

Methods based on fuzzy set concepts


  • Login